APPLIED PHYSICS LETTERS

VOLUME 82, NUMBER 3

20 JANUARY 2003

Formation of dispersions using ow focusing in microchannels

Shelley L. Anna, Nathalie Bontoux,a) and Howard A. Stoneb)
Division of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts 02138

Received 13 August 2002; accepted 21 November 2002 A ow-focusing geometry is integrated into a microuidic device and used to study drop formation in liquidliquid systems. A phase diagram illustrating the drop size as a function of ow rates and ow rate ratios of the two liquids includes one regime where drop size is comparable to orice width and a second regime where drop size is dictated by the diameter of a thin focused thread, so drops much smaller than the orice are formed. Both monodisperse and polydisperse emulsions can be produced. 2003 American Institute of Physics. DOI: 10.1063/1.1537519

Droplets of one uid in a second immiscible uid are useful in a wide range of applications, particularly when the droplet size and the size distribution can be prescribed on a micro- or nanoscale. As examples, many personal care products, foods,1 and products for topical delivery of drugs are emulsions, and nanoemulsions have been proposed for decontamination of surfaces infected in some way e.g., bacteria, bioterror agents, etc. .2 Similar emulsion structures are considered for organizing liquid-crystal droplets into optical devices.3,4 In this letter, we consider a ow-focusing conguration in a microuidic device for the formation of both monodisperse and polydisperse emulsions and we qualitatively illustrate aspects for controlling the drop size and distribution as the ow rates of the two liquid phases are varied. Emulsication methods are plentiful, but most involve mixing two liquids in bulk processes, and many use turbulence to enhance drop breakup. In these top-down approaches to emulsication, little control over the formation of individual droplets is available, and a broad distribution of sizes is typically produced.5 Alternatively, a bottom-up approach can be used for emulsication at the level of individual drops. Microuidic devices are ideal for thinking about this approach to microstructure formation.6 For example, Thorsen et al. formed emulsions in a microuidic device by colliding an oil stream and a water stream at a T-shaped junction.7 The resulting drops varied in size depending on the ow rate in each stream. Similar microuidic approaches to emulsication and two-phase ows have been described elsewhere.8 12 Alternatively, Ganan-Calvo and Gordillo produced highly monodisperse gas bubbles, less than 100 m in diameter,11 using a technique called capillary ow focusing: gas is forced out of a capillary tube into a bath of liquid, the tube is positioned above an orice, and the pressure-driven contraction ow of the external liquid through the orice focuses the gas into a thin jet, which subsequently breaks into equal-sized bubbles. In a separate experiment, this geometry was used to produce liquid droplets in air.13 Bubbles and drops formed by this ow-focusing technique are typically smaller than the upstream capillary tube and vary in size with the ow rates. In the present study, we report experiments using a owa b

focusing geometry, integrated into a planar microchannel design using soft lithography fabrication methods,14 to form liquid drops in a continuous phase of a second immiscible liquid. Such fabrication methods allow rapid production of an integrated microchannel prototype in essentially a single step.15 Using oil as the continuous phase and water as the dispersed phase, we observe a wide range of drop formation patterns, depending on the ow rates applied to each liquid inlet stream. We quantify the variation in size of the resulting water drops as a function of the oil ow rate, Q o , and the ratio of the internal water ow rate to the external oil ow rate, Q i /Q o . Both monodisperse and polydisperse patterns of drop formation occur, and the drop size can be either approximately independent of, or strongly dependent on, the ow rates, depending on the operating parameters chosen. Figure 1 shows the ow-focusing geometry implemented in a microuidic device: a liquid ows into the middle channel and a second immiscible liquid ows into the two outside channels. The two liquid phases are then forced to ow through a small orice that is located downstream of the three channels. The outer uid exerts pressure and viscous stresses that force the inner uid into a narrow thread, which then breaks inside or downstream of the orice. In the experiments reported here, the inner uid is distilled water and

Ecole Polytechnique, Palaiseau, Paris, France. Electronic mail: has@deas.harvard.edu

FIG. 1. Flow-focusing geometry implemented in a microuidic device. An orice is placed a distance H f 161 m downstream of three coaxial inlet streams. Water ows in the central channel, W i 197 m, and oil ows in the two outer channels, W o 278 m. The total width of the channel is W 963 m and the width of the orice is D 43.5 m. The thickness of the internal walls in the device is 105 m; this thickness is necessary in order to obtain a uniform seal between the glass cover slip and the polydimethylsiloxane PDMS . The uniform depth of the channels is 117 m. The design dimensions were slightly different than the measured values reported here since silicone oil swells the PDMS.

0003-6951/2003/82(3)/364/3/$20.00 364 2003 American Institute of Physics Downloaded 28 Sep 2010 to 141.212.59.68. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

Appl. Phys. Lett., Vol. 82, No. 3, 20 January 2003

Anna, Bontoux, and Stone

365

FIG. 2. Experimental images of drop breakup sequences occurring inside the ow-focusing orice. (a) Uniform-sized drops are formed without visible satellites; breakup occurs inside the orice. The time interval between images is 1000 s; Q o 8.3 10 5 mL/s and Q i /Q o 1/4. (b) A small satellite accompanies each large drop; breakup occurs at two corresponding locations inside the orice. The time interval between images is 166 s; Q o 4.2 10 4 mL/s and Q i /Q o 1/40.

the outer uid is silicone oil viscosity, 6 mPa s , which leads to water drops that form in a continuous phase of oil. Span 80 surfactant Sorbitan monooleate, Aldrich is dissolved in the oil phase at 0.67 wt %. The surfactant solution was prepared by mechanically mixing the two components for ap-

proximately 30 min and then ltering to eliminate aggregates and prevent clogging of the microchannel. The uids are introduced into the microchannel through exible tubing and the ow rate is controlled using separate syringe pumps for each uid. In the experiments reported here, the ow rate of the outer uid oil , Q o , is always greater than the ow rate of the inner uid water , Q i . Three different ow rate ratios are chosen, Q i /Q o 1/4, 1/40, and 1/400, where the oil ow rate given corresponds to the total ow rate for both oil inlet streams. For each Q i /Q o , oil ow rates spanning more than two orders of magnitude are chosen (4.2 10 5 mL/s Q o 8.3 10 3 mL/s). At each value of Q o and Q i , drop formation is visualized using an inverted microscope and a high-speed camera. In Fig. 2 a we show formation of a nearly monodisperse suspension of water droplets with diameter comparable to the orice width. Breakup occurs within the orice. In addition, drops may break within the orice such that one or more satellite droplets are formed in a regular and reproducible manner. Figure 2 b illustrates this kind of breakup process, which thus naturally forms a bidisperse suspension. We have conducted many experiments documenting the formation of two-phase liquidliquid dispersions in microchannels fabricated with the ow-focusing conguration. A phase diagram indicating the range of responses we have observed is shown in Fig. 3. We observe the formation of

FIG. 3. Phase diagram for drop formation in ow focusing. Each image represents the drop sizes and drop patterns that form at the specied value of Q o rows and Q i /Q o columns . We note that for these ows the Reynolds numbers R Q/( h), where is the kinematic viscosity of the uid and h is the height of the channel, of the oil (o) and water (i) are in the ranges 0.07 Ro 12 and 0.001 Ri 18, which are typically smaller than the values for the original ow-focusing studies see Refs. 11 and 12 . Downloaded 28 Sep 2010 to 141.212.59.68. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

366

Appl. Phys. Lett., Vol. 82, No. 3, 20 January 2003

Anna, Bontoux, and Stone

monodisperse droplets over a range of ow rates in which the drop size is approximately the size of the orice; see images a d , g , and m . In some cases coalescence occurs when droplets collide downstream of the orice; see images a c and m . Such collisions can occur at low ow rates when the droplets cannot travel downstream quickly enough. Coalescence can, in principle, be reduced by appropriate use of surfactants to stabilize the droplets. For other conditions, generally corresponding to higher oil ow rates and lower ratios of Q i /Q o images h j and n p , we observe bidisperse and polydisperse droplet distributions. Finally, the ability of the external uid to form narrow threads of the inner uid allows formation of monodisperse droplets with diameters much smaller than the orice width Figs. 3 k and 3 q , which is similar to the ow-focusing studies of Ganan-Calvo and co-workers. In conclusion, we have integrated a ow-focusing conguration into a microchannel fabricated with soft lithography techniques. The operating diagram of drop size as a function of ow rate and ow rate ratios illustrates regimes with both monodisperse and polydisperse droplets. The smallest droplets produced can be much smaller than the orice radius, in which case the drop size depends on the ow rates, and also there is a range of ow conditions where drops with diameters comparable to the orice width are formed independent of the ow rates. In fact, we have fabricated an orice approximately 10 m in width, for which the smallest drop sizes observed are in the range of hundreds of nanometers. Drop formation with this conguration offers interesting possibilities for designing emulsions. Finally, it is worth noting that in these small devices wetting issues can

be critical, as co-workers.16

discussed

recently

by

Dreyfus

and

The authors gratefully acknowledge support from the Unilever Corporation. The authors thank T. Jongen and colleagues for helpful conversations. One of the authors S.L.A. thanks A. Stroock and A. Winkleman for assistance in learning the soft lithography fabrication methods. The authors thank A. Shen, D. Link, and D. Weitz for helpful discussions.
C. Wibowo and K. M. Ng, AIChE J. 47, 2746 2001 . T. Hamouda, M. M. Hayes, Z. Cao, R. Tonda, K. Johnson, D. C. Wright, J. Brisker, and J. R. Baker, Jr., J. Infect. Dis. 180, 1939 1999 . 3 S. Matsumoto, Y. Sugiyama, S. Sakata, and T. Hayashi, Liq. Cryst. 27, 649 2000 . 4 G. De Filpo, J. Lanzo, F. P. Nicoletta, and G. Chidichimo, J. Appl. Phys. 84, 3581 1998 . 5 P. Becher, Emulsions: Theory and Practice Oxford University Press, Oxford, U.K., 2001 . 6 H. A. Stone and S. Kim, AIChE J. 47, 1250 2001 . 7 T. Thorsen, R. W. Roberts, F. H. Arnold, and S. R. Quake, Phys. Rev. Lett. 86, 4163 2001 . 8 V. Haverkamp, W. Ehrfeld, K. Gebauer, V. Hessel, H. Lowe, T. Richter, and C. Wille, Fresenius J. Anal. Chem. 364, 617 1999 . 9 M. W. Losey, M. A. Schmidt, and K. F. Jensen, Ind. Eng. Chem. Res. 40, 2555 2001 . 10 S. Sugiura, M. Nakajima, and M. Seki, Langmuir 18, 2854 2002 . 11 T. G. Mason and J. Bibette, Langmuir 13, 4600 1997 . 12 A. M. Ganan-Calvo and J. M. Gordillo, Phys. Rev. Lett. 87, 274501 2001 . 13 A. M. Ganan-Calvo, Phys. Rev. Lett. 80, 285 1998 . 14 J. C. McDonald, D. C. Duffy, J. R. Anderson, D. T. Chiu, H. Wu, O. J. A. Schueller, and G. M. Whitesides, Electrophoresis 21, 27 2000 . 15 G. M. Whitesides and A. D. Stroock, Phys. Today 54, 42 2001 . 16 R. Dreyfus, P. Tabeling, and H. Williams unpublished .
2 1

Downloaded 28 Sep 2010 to 141.212.59.68. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions